#!/usr/bin/env python3 """ ANALYSE DES RÉSULTATS D'OPTIMISATION MASSIVE Compile tous les fichiers et trouve les meilleures configs """ import json import os from datetime import datetime from collections import defaultdict import sys class MassiveResultsAnalyzer: def __init__(self, results_dir='optimization_results_massive'): self.results_dir = results_dir self.all_results = [] self.profitable = [] self.stats_by_leverage = defaultdict(lambda: {'total': 0, 'profitable': 0, 'best_pnl': -float('inf')}) def load_all_results(self): """Charge tous les fichiers de résultats""" print(f"📂 Lecture des résultats depuis: {self.results_dir}/") if not os.path.exists(self.results_dir): print(f"❌ Dossier non trouvé: {self.results_dir}") return files = [f for f in os.listdir(self.results_dir) if f.endswith('.json')] print(f"📄 {len(files)} fichiers trouvés\n") for filename in files: try: with open(os.path.join(self.results_dir, filename)) as f: result = json.load(f) self.all_results.append(result) # Stats lev = result['params']['leverage'] pnl = result['results']['total_pnl_pct'] pf = result['results']['profit_factor'] self.stats_by_leverage[lev]['total'] += 1 if pnl > 0 and pf > 1.0: self.profitable.append(result) self.stats_by_leverage[lev]['profitable'] += 1 if pnl > self.stats_by_leverage[lev]['best_pnl']: self.stats_by_leverage[lev]['best_pnl'] = pnl except Exception as e: print(f"⚠️ Erreur lecture {filename}: {e}") print(f"✅ {len(self.all_results)} configurations testées") print(f"✅ {len(self.profitable)} configurations profitables ({len(self.profitable)/len(self.all_results)*100:.1f}%)\n") def print_summary(self): """Affiche le résumé global""" print("="*100) print("📊 RÉSUMÉ GLOBAL - OPTIMISATION MASSIVE") print("="*100) print(f"Tests totaux: {len(self.all_results)}") print(f"Profitables: {len(self.profitable)} ({len(self.profitable)/len(self.all_results)*100:.1f}%)") print("="*100) def print_stats_by_leverage(self): """Stats par leverage""" print("\n📈 STATISTIQUES PAR LEVERAGE") print("="*100) for lev in sorted(self.stats_by_leverage.keys()): stats = self.stats_by_leverage[lev] total = stats['total'] profitable = stats['profitable'] rate = profitable / total * 100 if total > 0 else 0 best_pnl = stats['best_pnl'] if stats['best_pnl'] > -float('inf') else 0 print(f"{lev:3}x | Tests: {total:4} | Profitables: {profitable:3} ({rate:5.1f}%) | Meilleur PnL: {best_pnl:+7.1f}%") print("="*100) def print_top_configs(self, n=50): """Affiche les TOP N configs""" if not self.profitable: print("\n❌ Aucune configuration profitable trouvée!") return print(f"\n🏆 TOP {n} CONFIGURATIONS PROFITABLES") print("="*100) # Trier par PnL décroissant top = sorted(self.profitable, key=lambda x: x['results']['total_pnl_pct'], reverse=True)[:n] for idx, result in enumerate(top, 1): params = result['params'] res = result['results'] print(f"\n#{idx}. PnL: {res['total_pnl_pct']:+6.1f}% | " f"PF: {res['profit_factor']:5.2f} | " f"WR: {res['win_rate']:5.1f}% | " f"Trades: {res['total_trades']:3}") print(f" Leverage: {params['leverage']:3}x | " f"SL: {params['sl_atr_mult']:4.2f}x ATR | " f"TP: {params['tp_atr_mult']:4.2f}x ATR | " f"Ratio: {params['tp_atr_mult']/params['sl_atr_mult']:4.1f}:1") print(f" Min Conf: {params['min_confidence']:2}% | " f"Position: {params['position_pct']:2}% | " f"Final: ${res['final_capital']:7.2f}") print(f" Avg Win: ${res['avg_win']:6.2f} | " f"Avg Loss: ${res['avg_loss']:7.2f}") print("="*100) def save_report(self, filename='massive_optimization_report.txt'): """Sauvegarde un rapport complet""" with open(filename, 'w') as f: f.write(f"RAPPORT D'OPTIMISATION MASSIVE\n") f.write(f"Généré le: {datetime.now()}\n") f.write("="*100 + "\n\n") f.write(f"RÉSUMÉ:\n") f.write(f" Tests totaux: {len(self.all_results)}\n") f.write(f" Profitables: {len(self.profitable)} ({len(self.profitable)/len(self.all_results)*100:.1f}%)\n\n") f.write("STATS PAR LEVERAGE:\n") for lev in sorted(self.stats_by_leverage.keys()): stats = self.stats_by_leverage[lev] total = stats['total'] profitable = stats['profitable'] rate = profitable / total * 100 if total > 0 else 0 best_pnl = stats['best_pnl'] if stats['best_pnl'] > -float('inf') else 0 f.write(f" {lev:3}x | Tests: {total:4} | Profitables: {profitable:3} ({rate:5.1f}%) | Meilleur: {best_pnl:+7.1f}%\n") f.write("\n" + "="*100 + "\n") f.write("TOP 50 CONFIGURATIONS\n") f.write("="*100 + "\n\n") top = sorted(self.profitable, key=lambda x: x['results']['total_pnl_pct'], reverse=True)[:50] for idx, result in enumerate(top, 1): params = result['params'] res = result['results'] f.write(f"#{idx}. PnL: {res['total_pnl_pct']:+6.1f}% | PF: {res['profit_factor']:5.2f} | WR: {res['win_rate']:5.1f}% | Trades: {res['total_trades']:3}\n") f.write(f" Leverage: {params['leverage']:3}x | SL: {params['sl_atr_mult']:4.2f}x ATR | TP: {params['tp_atr_mult']:4.2f}x ATR | Ratio: {params['tp_atr_mult']/params['sl_atr_mult']:4.1f}:1\n") f.write(f" Min Conf: {params['min_confidence']:2}% | Position: {params['position_pct']:2}% | Final: ${res['final_capital']:7.2f}\n\n") print(f"\n💾 Rapport sauvegardé: {filename}") def run(self): """Execute l'analyse complète""" self.load_all_results() if len(self.all_results) == 0: print("❌ Aucun résultat à analyser!") return self.print_summary() self.print_stats_by_leverage() self.print_top_configs(n=50) self.save_report() if __name__ == '__main__': analyzer = MassiveResultsAnalyzer() analyzer.run()